Renal Cell Carcinoma-Infiltrating CD3low Vγ9Vδ1 T Cells Represent Potentially Novel Anti-Tumor Immune Players
, Minyong Kang 4, Yun Shin Chung 2, Won Joon Oh 2, Seon-Yong Yeom 4, Woong-Yang Park 5, Tae Jin Kim 2,* and Seong Il Seo 4,*Round 1
Reviewer 1 Report
This study focused on gamma-delta (γδ) T cell subsets and mechanisms in RCC. Authors found CD3low γδ T cells was significantly higher in freshly resected tumor specimens than in peripheral blood. Of these intra-tumoral CD3low γδ T cells, CD3low Vγ9Vδ1 T cells differentiated effector memory phenotype and have the cytotoxic activity against ACHN RCC cell line. They concluded that CD3low Vγ9Vδ1 T cells may represent an excellent candidate for adoptive immunotherapy in high-risk patients with locally advanced RCC. This study may have limited interest to the clinician and researcher, because the clinical importance of CD3low Vγ9Vδ1 T cells is still unclear. Author must address the following criticisms listed as below.
#1 The association with proportion of CD3low Vγ9Vδ1 T cells and clinical characteristics
As I mentioned above, the clinical importance of CD3low Vγ9Vδ1 T cells is unclear. Authors should compare each CD3low Vγ9Vδ1 T cells population in clinical characteristics, such as TNM stage, grade, sex and age.
#2 Small sample size and inaccurate data
Sample size is too small to discuss. Particularly, only 6-7 sample use in Fig3 and Fig4, and not describe which sample selected.
Furthermore, the proportion values of FACS data and dot graph are different in Fig2B (CD3high γδ TCR+ TIL) and Fig3B (PBL).
#3 The different results according to cell lines
In this study, the cytotoxic activity of CD3low Vγ9Vδ1 T cells was detected only ACHN cell, but not Caki-1 cell. The authors should discuss a reasonable theory as to why / how occurred the different results according to cell lines.
#4 The association with proportion of CD3low Vγ9Vδ1 T cells and therapeutic outcome or prognosis
It will be more useful if authors can discuss the relation between the level of CD3low Vγ9Vδ1 T cells and therapeutic outcome, such as TKI and mTOR inhibitors (using TCGA data).
Author Response
Please see the attachment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
The Authors have isolated and characterized the tumor infiltrating CD3lowVg9Vd1T cells in RCC tumors; and performed bioinformatic analysis to support the anti-tumorigenic roles of these cells.
They have performed in vitro analysis to study the cancer-cell lysing properties of these cells. The results from this study correlate well with the existing literature on the potential anti-tumor effector functions of gdT cells. [g-gamma, d-delta].
However, few points need to be addressed before the conclusion, as stated in the manuscript, can be accepted.
- Almost all 20 tumors analyzed were of clear cell phenotype. The in vitro experiments lack 786-O cells or VHL-null clear cell (Caki-1 is VHL positive as the authors are aware of). Inclusion of VHL-null cells becomes important, as tumor infiltrating Vg9Vd1T cells are shown to express high PD-1, and PD-L1 is very high on VHL-null RCC cells. Whether Vg9Vd1T cells still retain cytotoxic effects on VHL-null cells remains unanswered.
- Since gdT cells are not dependent on MHC, a co-culture of PBL Vg9Vd2T cells with different RCC cell lines and analyzing the immunophenotypic changes can reveal a possible mechanism for the influence of RCC TME on the development of tumor infiltrating Vg9Vd1T cells: the authors can perform these experiments if feasible.
- In the co-culture assays (like Fig-4A), anti-PD-1 antibody should be added to assess whether Vg9Vd1T cells can still retain cytotoxic functions; and this can give a glimpse of the potential effects of the use of anti-PD1 treatments on Vg9Vd1T cells.
- If adoptive transfer of these cells becomes feasible, mobilization of Vg9Vd2T cells to RCC tumors becomes important; and CXCL9 in TME and CXCR3 on T cells promote infiltration. In this context, have the authors checked the CXCR3 gene expression on Vg9Vd1T cells?
Performing the above described experiments will support the conclusions stated in the manuscript.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
The proportion values of FACS data (left) and dot graph (right) are different in Fig2B (CD3high γδ TCR+ TIL) and Fig3B (PBL).
Authors should revise this.
Reviewer 2 Report
Dear Authors,
From the histogram presented for CXCR3 analysis, I can see that d1T and d2T cells express CXCR3 (from the shift compared to isotype control). Yet, in the response it was stated CXCR3 was NOT expressed. Please clarify.
